EVOLUTIONARY MACHINE LEARNING DRIVEN BIG DATA ANALYSIS AND PROCESSING FOR INDUSTRIAL INTERNET

Abstract

The Industrial Internet is based on the network, the platform is the core, and the security is the guarantee. The Industrial Internet connects all industry elements and the entire industry chain through the large-scale network infrastructure, collects and analyzes industry data in real time, and forms a new application model for a new generation of information communication. With the rapid development of industrial Internet technology, the scale of industrial Internet data will become larger and larger, and the data dimension will become higher and higher. How to efficiently use cluster analysis for industrial Internet big data mining is an urgent problem that needs to be solved. This paper proposes an improved differential evolution particle swarm algorithm for industrial Internet big data clustering analysis. Differential Evolution (DE) strategy can improve the problem that the particle swarm optimization (PSO) algorithm tends to fall into local optimum in the later stage as the number of iterations increases. Considering the influence of the randomness of the arrangement order of the cluster center vectors among the individuals on the learning and updating among individuals, this paper designs a method of adaptively adjusting the arrangement order of the cluster center vectors to optimize the cluster center vector with maximum similarity among individuals. In order to effectively evaluate our method, both industrial and non-industrial datasets are selected. The experimental results verify the feasibility and effectiveness of the proposed algorithm.